Intro to Marketing Research: Data Analysis - Descriptive Measures of Dispersion, Range Variance Standard Deviation Coefficient of Variation Interquartile Range

Measures of Dispersion

What It Is

Measures of dispersion are statistical tools used to quantify the spread or variability of a dataset. They help researchers and marketers understand how data points are distributed around the mean value. A classic example is the study by Galton (1886) on the relationship between the heights of parents and their children. Galton used measures of dispersion to show that the heights of children were not a simple average of their parents' heights, but rather a normal distribution with a significant spread. This matters for marketing decision-making as it helps in understanding customer behavior, preferences, and expectations.

Key Terms & Concepts

• Range: The difference between the highest and lowest values in a dataset. It is a simple measure of dispersion but can be affected by outliers. (Example: A company's sales data shows a range of $100 to $10,000, indicating a significant variation in sales.)
• Variance: A measure of the average squared differences from the mean. It is calculated as the sum of squared deviations divided by the number of observations. (Formula: ?² = ?(xi - ?)² / (n - 1), where ?² is the variance, xi is each data point,-is the mean, and n is the sample size.)
• Standard Deviation: The square root of the variance. It is a more intuitive measure of dispersion as it is expressed in the same units as the data. (Formula:-= ²)
• Coefficient of Variation: A relative measure of dispersion that expresses the standard deviation as a percentage of the mean. It is useful for comparing the variability of different datasets. (Formula: CV =-/-× 100)
• Interquartile Range (IQR): The difference between the 75th percentile (Q3) and the 25th percentile (Q1). It is a robust measure of dispersion that is less affected by outliers. (Example: A company's customer satisfaction data shows an IQR of 20, indicating a moderate variation in satisfaction levels.)
• Skewness: A measure of the asymmetry of a distribution. It can be positive (right-skewed), negative (left-skewed), or zero (symmetric). (Example: A company's sales data shows a positive skewness, indicating that most sales are concentrated at the lower end of the range.)
• Kurtosis: A measure of the "tailedness" of a distribution. It can be platykurtic (flat), leptokurtic (peaked), or mesokurtic (normal). (Example: A company's customer satisfaction data shows a leptokurtic distribution, indicating that most customers are highly satisfied or highly dissatisfied.)
• Outlier: A data point that is significantly different from the rest of the dataset. It can affect measures of dispersion and should be handled carefully. (Example: A company's sales data shows an outlier of $100,000, which is significantly higher than the rest of the data.)
• Robustness: A measure of a statistical method's ability to withstand the presence of outliers or other data anomalies. (Example: The IQR is a robust measure of dispersion as it is less affected by outliers.)
• Sensitivity: A measure of a statistical method's ability to detect small changes in the data. (Example: The coefficient of variation is a sensitive measure of dispersion as it can detect small changes in the standard deviation.)

Common Misunderstandings

1. Misunderstanding: The range is a good measure of dispersion for all datasets. Correction: The range is sensitive to outliers and should be used with caution. A more robust measure like the IQR is often preferred.
2. Misunderstanding: The standard deviation is always a good measure of dispersion. Correction: The standard deviation is sensitive to outliers and may not be the best choice for all datasets. The IQR or coefficient of variation may be more suitable.
3. Misunderstanding: The coefficient of variation is only used for comparing datasets with the same units. Correction: The coefficient of variation can be used to compare datasets with different units, as it expresses the standard deviation as a percentage of the mean.

Quick Application / Identification

Scenario: A company wants to understand the variation in customer satisfaction levels across different regions. The data shows a mean satisfaction level of 80 with a standard deviation of 10. What is the coefficient of variation?

Answer: 12.5% (correct answer: CV =-/-× 100 = 10 / 80 × 100 = 12.5%)

Explanation: The coefficient of variation helps the company understand the relative variation in customer satisfaction levels across different regions.

Last-Minute Revision

• The range is sensitive to outliers
• The variance is calculated as the sum of squared deviations divided by the sample size (?² = ?(xi - ?)² / (n - 1))
• The standard deviation is the square root of the variance (? = ²)
• The coefficient of variation is a relative measure of dispersion (CV =-/-× 100)
• The IQR is a robust measure of dispersion (IQR = Q3 - Q1)
• Skewness measures the asymmetry of a distribution
• Kurtosis measures the "tailedness" of a distribution
• Outliers can affect measures of dispersion
• Robustness is a measure of a statistical method's ability to withstand outliers
• Sensitivity is a measure of a statistical method's ability to detect small changes in the data